CRule: Category-Aware Symbolic Multi-Hop Reasoning on Knowledge Graphs

CORC > 自动化研究所 > 中国科学院自动化研究所 > 舆论大数据科学与技术应用联合实验室

	CRule: Category-Aware Symbolic Multi-Hop Reasoning on Knowledge Graphs
	Wang, Zikang; Li, Linjing; Li, Jinlin; Zhao, Pengfei; Zeng, Daniel
刊名	IEEE Intelligent Systems
	2023
页码	1-9
英文摘要	Multi-hop reasoning is essential in knowledge graph (KG) research and applications. Current methods rely on specific KG entities, while human cognition operates at a more abstract level. This paper proposes a Category-aware Rule-based (CRule) approach for symbolic multi-hop reasoning. Specifically, given a KG, CRule first categorizes entities and constructs a category-aware KG, then uses rules retrieved from the categorized KG to perform multi-hop reasoning on the original KG. Experiments on five datasets show that CRule is simple, effective, and combines the advantages of symbolic and neural network methods. It overcomes symbolic reasoning’s complexity limitations, can perform reasoning on KGs of more than 300k edges, and can be three times more efficient than neural network models.
语种	英语
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/52343]
专题	舆论大数据科学与技术应用联合实验室
作者单位	Institute of Automation, Chinese Academy of Sciences
推荐引用方式 GB/T 7714	Wang, Zikang,Li, Linjing,Li, Jinlin,et al. CRule: Category-Aware Symbolic Multi-Hop Reasoning on Knowledge Graphs[J]. IEEE Intelligent Systems,2023:1-9.
APA	Wang, Zikang,Li, Linjing,Li, Jinlin,Zhao, Pengfei,&Zeng, Daniel.(2023).CRule: Category-Aware Symbolic Multi-Hop Reasoning on Knowledge Graphs.IEEE Intelligent Systems,1-9.
MLA	Wang, Zikang,et al."CRule: Category-Aware Symbolic Multi-Hop Reasoning on Knowledge Graphs".IEEE Intelligent Systems (2023):1-9.